Make.com Zero to Hero 2026 – 2.4 AI Basics in Make · published 9 November 2026
Play at 1:48 Watch on YouTube Make.com
What is a prompt, a model, and a token when I use AI in Make.com?
The prompt is the instruction in plain words. The model is which AI answers it. A token is the small piece of text the model actually counts, about four characters, or roughly three quarters of a word.
A text model takes words and gives words back. That is the one I use most: captions, post ideas, hashtags, and replies to comments. An image model takes a description, such as a bright gym at sunrise, and returns a new picture. A video model returns a short clip. The instruction goes in and a result comes back. Only the kind of result changes. A good prompt says what it is about, the tone, and how long it should be. A clearer prompt gets a better result. A small model is fast and cheap. A big model is smarter and costs more. For social posts I start small and I only go bigger if the result is not good enough. I pay for the tokens I send in and for the tokens the answer writes. The answer usually costs more per token. Text prices are per 1 million tokens, and one caption is a tiny fraction of that. Images and videos are usually priced per picture or per second. One hundred tokens is about 75 words.